It's not rationalization at all. It's "making decisions based on presented evidence".
A good example of this is Sylow's DNC gearing thread, which is a really great resource full of useful information that begins with the explicit premise that you'll always have a healer and thus won't have to waltz, which is a totally reasonable caveat and is necessary for the purpose of the thread (since I can't even quantify how much I'll have to heal so I certainly can't expect Sylow to do it for me) but only works as an actual handwave if I totally ignore everything I actually use my DNC for.
Lv.99 MNK TP Set
|
lv.99 MNK TP set
You basically just wrote one of the definitions of "rationalizing".
That said, I'm rationalizing too, except mine is valid more often and isn't being treated as an absolute. Valefor.Prothescar said: » You basically just wrote one of the definitions of "rationalizing". A definition popular with people who don't like complex cost/benefit analyses, yeah. Quote: isn't being treated as an absolute. Well, I hope you'll forgive me if I'm not going to argue that pchan isn't generally pretty bad at making his case D: I've probably only seen him express himself properly one time
We're all just looking at the numbers in different ways. And the mob HP and what you expect from a parse does make a difference. This is why:
Let's assume that you added 4% quad to your sets and expect (in order to gain your advantage over other sets, you need to quad at least once in 25 hits). The equation for this is: Many hits: 100*(1-.96^x) while x is the number of hits that you would need to perform. This gives you the probability that you'd hit at least 1 quad attack. At only 25 hits, this happens to be 63% chance. That would mean that 63% of the time, you will win the parse because of that critical quad attack, and the other 36% chance you'd lose because no procs occured. Few Hits If we decided that a mob only had 10 hits required, then Pchan's assumptions are correct because 100*(1-.96^10) = 33% chance of winning with a quad and 66% chance to win with pure consistency. The entire model is turned on its head when we have to consider the type of mobs we're killing and what type of parsing the player wants to get out of it. Relying on 1 quad attack round to put a particular parse to your favor would be dumb if you could only count on it happening in 5% of each of your fight (winning 5 parses out of 100). But on the otherhand, if there was more hits, the probability of that needed quad(or multi) occuring in order to give you that advantage appears. tldr: 1. If a mob has too few HP, and there are few mobs and we care about how many of those mobs we outparse others in, the "Pchan consistancy set" is better. It is inherent in the proc rate of the gear you're adding how many hits are considered per kill. 2. Over long term parses, the damage will reflect whatever the DPS spreadsheets show. I think: Whether you win parses DOES depend on the amount of variance that Pchan describes depending on the mob. He's arguing that overkill damage on one mob is a win, but if none of the needed procs occur, you will lose. However, I think he's underestimating the amount of hits that a MNK actually performs and there was no indication of what mob he's trying to parse. One model is truncated (Pchan). One model is continuous. Whichever set you decide on depends on what you value and everyone values different things. It's a shame, because there's sometimes a worthwhile point to be made underneath it* but I really do not have the time or energy to be the guy that consistently leaps to his defense.
*and to give him the benefit of the doubt, he's usually fighting an uphill battle against conventional wisdom to begin with and the fact that English isn't his first language seriously does not help Bismarck.Keityan said: » However, I think he's underestimating the amount of hits that a MNK actually performs and there was no indication of what mob he's trying to parse.
[+]
Phoenix.Suji said: » Bismarck.Keityan said: » However, I think he's underestimating the amount of hits that a MNK actually performs and there was no indication of what mob he's trying to parse. Give him a LITTLE credit; he's been the first to acknowledge that low-man ADL is basically over when the PD/Embrava nerf goes live, so he's probably been looking forward at least a little bit here. ScaevolaBahamut said: » Phoenix.Suji said: » Bismarck.Keityan said: » However, I think he's underestimating the amount of hits that a MNK actually performs and there was no indication of what mob he's trying to parse. Give him a LITTLE credit; he's been the first to acknowledge that low-man ADL is basically over when the PD/Embrava nerf goes live, so he's probably been looking forward at least a little bit here. Edit: And to be fair, I could see the argument for consistency being stronger if one is truly optimizing specifically for ADL. The bottom line is that whatever you are fighting will fall into one of two categories:
1) Enough hit points where in a high percetage of cases multiattack will occur near its listed value(i.e. ADL, VW, etc) 2) A large number of mobs where eventually multiattack will approach its listed value over time(NNI, salvage II, etc). There is no 3) Fighting one mob that has 500 hit points then logging out for the night.
[+]
Siren.Barber said: » The bottom line is that whatever you are fighting will fall into one of two categories: 1) Enough hit points where in a high percetage of cases multiattack will occur near its listed value(i.e. ADL, VW, etc) 2) A large number of mobs where eventually multiattack will approach its listed value over time(NNI, salvage II, etc). There is no 3) Fighting one mob that has 500 hit points then logging out for the night. bollocks. If you are running an event with a bunch of mobs with 6k health and you always know your WS will do 2500, for instance, that consistency is EXTREMELY RELEVANT and you will fight accordingly. Except this is the "Monk TP thread" so getting the mob to where it has 2500 left is the discussion....not how much the ws will do. That is a different argument....and a different thread.
edit: unless you are arguing you need to know exactly how long it will take to do 3500 damage so you can ws. Even then it doesn't matter, you still have to account for 3500 white damage which over time will favor the higher dot setup if given enough mobs. For reference, under the conditions I listed previously, here are the 10 best sets (first number is the average DPS, then the gear list). The third one seems to be the the least dependant to variance.
Code 752.416280,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,nomkahpa,rajas,epona,atheling,twilight,byakko,usukane, 751.501799,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,tenryu +1,rajas,epona,atheling,twilight,byakko,usukane, 749.972758,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,nomkahpa,rajas,epona,atheling,black belt,tantra +2,thaumas, 749.359653,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,tenryu +1,rajas,epona,atheling,black belt,tantra +2,thaumas, 749.231037,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,twilight,tantra +2,thaumas, 748.440219,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,twilight,byakko,usukane, 748.353914,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,windbuffet,tantra +2,usukane, 745.666462,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,nomkahpa,rajas,epona,atheling,black belt,byakko,usukane, 744.837042,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,nomkahpa,rajas,epona,atheling,twilight,tantra +2,usukane, 744.811300,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,nomkahpa,rajas,epona,atheling,twilight,byakko,usukane, I'm still using 2.25 ratio cap, even match infinite HP mobs and augmented byakko... Why should I believe anything your simulation spits out without having the code to analyze?
[+]
why the *** are you using a 2.25 ratio cap for MNK
[+]
He's looking toward the 1H adjustment.
[+]
Fenrir.Sylow said: » He's looking toward the 1H adjustment. Bad idea given the response the other day about H2H being treated independently. If H2H gets left out of that adjustment I will be supremely annoyed. Would anyone be able to give me a link to one of the spreadsheets for calculating this stuff? I am interested in checking how well this set compares to the widely accepted optimum:
I chose these pieces in an effort to keep ~25% haste and maintain a high level of multi-attack without extravagant gear like Tenryu Tekko +1. Comparison with the set with Tenryu Tekko +1, Windbuffet belt, Thew Bomblet, and Usukane Sune-Ate +1: Double Attack +5% Triple Attack -2% Quadruple Attack -1% Accuracy -4 Attack -18 STR -2 Store TP -9 Increased kick attack damage Small chance for two kick attacks in a round Here are the 10 best TP sets under the following assumptions
boost DEX / minuets x2 (relic) / marchesx2 (relic) / haste / no berserk / dia2 Typical target : level 99, 470 eva, 560 def, 95 agi, 95 vit. Impetus up : Code 755.183444,verethragna 99,demonry,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,windbuffet,tantra +2,usukane, 753.701523,verethragna 99,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,windbuffet,tantra +2,usukane, 752.270723,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,nomkahpa,rajas,epona,atheling,twilight,byakko,usukane, 751.449640,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,tenryu +1,rajas,epona,atheling,twilight,byakko,usukane, 750.786937,verethragna 99,hagneia,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,windbuffet,tantra +2,usukane, 750.644228,verethragna 99,hagneia,usukane,rancor,brutal,ghillie +1,thaumas,tenryu +1,rajas,epona,atheling,windbuffet,tantra +2,usukane, 750.360085,spharai 99 afterglow,demonry,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,windbuffet,tantra +2,usukane, 750.134580,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,nomkahpa,rajas,epona,atheling,black belt,tantra +2,thaumas, 749.773310,spharai 99 afterglow,hagneia,usukane,rancor,brutal,moonshade att/reg,tantra +2,nomkahpa,rajas,epona,atheling,twilight,byakko,usukane, 749.585530,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,tantra +2,tenryu +1,rajas,epona,atheling,black belt,tantra +2,thaumas, Impetus down Code 700.580464,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,twilight,tantra +2,thaumas, 699.729123,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,windbuffet,tantra +2,usukane, 697.046095,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,twilight,byakko,usukane, 695.468581,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,black belt,tantra +2,thaumas, 694.796600,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,nomkahpa,rajas,epona,atheling,black belt,tantra +2,thaumas, 694.239489,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,nomkahpa,rajas,epona,atheling,twilight,tantra +2,thaumas, 693.747117,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,tenryu +1,rajas,epona,atheling,twilight,tantra +2,usukane, 693.447577,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,nomkahpa,rajas,epona,atheling,twilight,byakko,usukane, 693.425134,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,seiryu,rajas,epona,atheling,black belt,tantra +2,thaumas, 692.393878,spharai 99 afterglow,potestas bomblet,usukane,rancor,brutal,moonshade att/reg,thaumas,seiryu,rajas,epona,atheling,twilight,abatteur,usukane, Those only account for average value. Vere is done with AM1 so it's slightly overestimated since you can't have it 100% in small fights. pchan said: » Vere is done with AM1 so it's slightly overestimated since you can't have it 100% in small fights. Oh nvm, you're talking about best dps against fodder mobs. Is that particularly relevant? This comparison is for normal players not retards though.
Your target's stats are bizarre.
[+]
pchan said: » This comparison is for normal players not retards though. Then shouldn't you be excluding yourself? Your numbers are notorious for being skewed in your favor, and this is no exception since you're putting an inhibitor on the Vere. Most situations that Monks are going to be in nowadays will allow them to have the time to build their TP to 300% to start the fight with AM3 available, or have the temp items available to them to hit 300% TP quick. Likewise, it's against fodder mobs, which, as Peldin said, isn't particularly relevant since it's only providing part of the picture as far as the best TP sets and, knowing you, was selected intentionally just so that your gear set of choice would come out as being better. By most situations you are probably referring to VoidWatch ? There will be no variation against stronger mobs for obviosu reasons, but I guess it's not obvios to everyone ? Attack is uncapped yo.
Idk, VW, Odin, ADL all give you a chance to build to 300 before popping, in Legion I think embrava is staying untouched(? unsure)..so Salvage and Meeble are left where you can't?
Anyone care to comment on my post? I'd appreciate good advice.
pchan said: » This comparison is for normal players not retards though. I only asked if level 99 targets was a relevant target for determining top dps gear sets. Shouldn't we be using target levels that are more relevant to what we actually do? Valefor.Sehachan said: » Idk, VW, Odin, ADL all give you a chance to build to 300 before popping, in Legion I think embrava is staying untouched(? unsure)..so Salvage and Meeble are left where you can't? With Salvage2, many of the fodder and all the bosses are over level 99 as well. So anyway, that was the only reason I asked about the relevancy of evaluating level 99 targets. They aren't typical for parses that we actual care about. But hey, I guess it's for normal players, not retards that fight high level mobs. Fenrir.Sylow said: » Why should I believe anything your simulation spits out without having the code to analyze? You are too dumb to realize that I'm always right but ok let's do eet. I wonder if you 'll go through the 7000 lines of the most advanced DPS simulator. https://docs.google.com/file/d/0B7wbcilaHsTecldhZm94a1gtZnM/edit |
All FFXIV and FFXI content and images © 2002-2025 SQUARE ENIX CO., LTD.
FINAL FANTASY is a registered trademark of Square Enix Co., Ltd.